viz: fix BGR/RGB channel swap in readCloud PLY, add PLY I/O and traje…

[rtkartista]

Summary

Fix a long-standing BGR/RGB channel swap bug in cv::viz::readCloud() for PLY files,
and add a test suite for PLY cloud I/O and trajectory read/write roundtrips.

Bug fixed

readCloud() returns colors in RGB order (as stored in PLY spec) rather than
BGR (OpenCV convention). writeCloud() correctly converts BGR→RGB on write,
but the inverse conversion was missing on read. Every color channel 0 and 2
were silently swapped for all callers.

Fix: swap channels 0↔2 in-place after sink->Write() in vizcore.cpp.

Tests added (modules/viz/test/test_ply_io.cpp)

PLY cloud I/O

• ASCII roundtrip — positions preserved within 1e-4 tolerance
• Binary roundtrip — positions bit-exact for float32
• Binary roundtrip with colors — regression test for the BGR/RGB fix
• Large cloud (500k pts) binary perf smoke test (<2s read+write)
• Empty cloud — no crash

Trajectory I/O

• Affine3d write/read roundtrip — 1e-9 tolerance
• Affine3f write/read roundtrip — 1e-5 tolerance
• Non-zero start index preserved correctly
• readTrajectory stops cleanly on missing file
• Empty range returns empty result

Secondary issue noted

readTrajectory() with an untyped std::vector OutputArray and zero results
calls Mat(traj).convertTo(_traj, _traj.depth()) where _traj.depth() returns
-1, causing undefined behavior. Guard with if (!traj.empty()) or assign a
typed empty array. Tracked separately.

Related

Contributes toward the SLAM module I/O test coverage noted in the
GSoC 2026 / 2026 wiki.

Pull Request Readiness Checklist

See details at https://github.com/opencv/opencv/wiki/How_to_contribute#making-a-good-pull-request

• I agree to contribute to the project under Apache 2 License.
• To the best of my knowledge, the proposed patch is not based on a code under GPL or another license that is incompatible with OpenCV
• The PR is proposed to the proper branch
• There is a reference to the original bug report and related work
• There is accuracy test, performance test and test data in opencv_extra repository, if applicable
  Patch to opencv_extra has the same branch name.
• The feature is well documented and sample code can be built with the project CMake